Journal of Experimental Botany, Vol. 51, No. 352, pp. 1945±1947, November 2000
A novel DNA-binding protein associated with DNA polymerase-a in pea stimulates polymerase activity on infrequently primed templates John A. Bryant1,3, David C. Brice1, Paul N. Fitchett1 and Louise E. Anderson2 1 2
School of Biological Sciences, University of Exeter, Exeter EX4 4QG, UK Department of Biological Sciences, University of Illinois, Chicago, IL 60607-7060, USA
Received 1 March 2000; Accepted 13 July 2000
Abstract A 42 kDa DNA-binding protein is associated with DNA polymerase-a-primase in pea (Pisum sativum). In a previous publication it was shown that the protein has strong preference for ds±ss junctions in DNA, including the cohesive termini generated by restriction endonucleases. In this paper it is shown that when the DNA-binding protein is added back to polymerase-primase, the protein stimulates the activity of the polymerase. The stimulation is particularly marked when M13 DNA, primed with a single sequencing primer or primed with oligoribonucleotides by the polymerase's associated primase activity, is used as a template. The stimulation of polymerase activity is not caused by an increase in processivity. These data lead to the suggestion that the 42 kDa DNA-binding protein is a primer-recognition protein. Key words: DNA-binding protein, DNA polymerase, pea, Pisum sativum, primer, primer-recognition protein.
Introduction Current models for nuclear DNA replication in eukaryotes involve three different DNA polymerases, a, d and e (Waga and Stillman, 1998). Of these, only polymerase-a is associated with a primase activity and is, therefore, the only nuclear DNA polymerase that is able to initiate new DNA molecules (Waga and Stillman, 1994). This enzyme exists as part of a multi-protein complex in eukaryotes as diverse as fungi, invertebrates, vertebrates, and plants (Bryant et al., 1992). In previous work on characterization of the polymerase-a-primase complex in pea (Pisum 3
sativum) it was shown that the complex contains at least two DNA-binding proteins (Bryant et al., 1992; Al-Rashdi and Bryant, 1994). One of these, a 42 kDa polypeptide, has been puri®ed and partially characterized, showing that its DNA-binding properties are very unusual (Al-Rashdi and Bryant, 1994; Burton et al., 1997). The main features of its binding activity are that it binds only very weakly or not at all to completely ss or to completely ds DNA but that it does bind to ds±ss junctions with an especially strong af®nity for the cohesive termini generated by restriction endonucleases. These properties, taken with the size of the protein (42 kDa) led to it being compared with a primer recognition protein which is associated with DNA polymerase-a in human cells (Burton et al., 1997). In this paper it is shown that the pea DNA-binding protein stimulates the activity of DNA polymerase-a in a manner consistent with primer-recognition activity.
Materials and methods Plant material Shoot apices of etiolated 5±7-d-old pea seedlings were used as the source of DNA-polymerase-a-primase (Bryant et al., 1992).
Protein purification Pea DNA polymerase-a-primase was prepared and assayed as described earlier (Bryant et al., 1992) except that the gradient used to elute the proteins from phosphocellulose was 0.0 M to 0.6 M K-phosphate rather than 0.0 M to 1.0 M. In the latter gradient system the 42 kDa DNA-binding protein elutes with or very close to the polymerase-primase peak (Bryant et al., 1992; Al-Rashdi and Bryant, 1994). Elution of the proteins from the phosphocellulose column in the shallower (0.0 M to 0.6 M) phosphate gradient led to a much better separation of the
To whom correspondence should be addressed. Fax: q44 1392 264668. E-mail: [email protected]
ß Society for Experimental Biology 2000
Bryant et al.
DNA-binding protein from the polymerase. Further puri®cation on heparin-agarose (Al-Rashdi and Bryant, 1994) yielded one DNA-binding activity of 42 kDa. The partial breakdown product of 36 kDa that was reported in earlier papers (Al-Rashdi and Bryant, 1994; Burton et al., 1997) was not seen in preparations made by this slightly modi®ed procedure. DNA polymerase assays DNA polymerase assays were carried out as described previously (Bryant et al., 1992), using 16 mg DNA polymerase per assay. In `add-back' assays 1.6 mg of primer-recognition protein was added per assay. In all assays, three or four replicates were done for each time point and each processed sample was `counted' in the scintillation counter for 10 min.
Processivity assays For processivity assays, M13 primers were end-labelled with 32P and then used to prime M13 DNA prior to assay of DNA polymerase. The labelled reaction products resulting from elongation of the primers were separated on polyacrylamide gels which were then autoradiographed (Daniel et al., 1985).
Results Add-back experiments
DNA polymerase-a-primase was assayed in the absence or presence of the 42 kDa DNA-binding protein with three different template-primer systems, namely (A) activated calf-thymus DNA (the standard assay system), (B) M13 DNA primed with a single sequencing primer and (C) M13 primed by the polymerase-primase itself. For this latter assay, non-radioactive ribonucleoside59-triphosphates were added to the assay mix (Bryant et al., 1992). In the activated calf-thymus DNA system (A) the primers are very frequently spaced on the template: the primers are of varying length and are separated by gaps of 30±70 nucleotides of single-stranded template (Pritchard et al., 1983). For primed M13 (B) the primer is a single sequence-speci®c oligonucleotide per c. 6400 nucleotides of single-stranded template. Thus, it is possible to compare infrequently primed (M13) template with very frequently primed (activated DNA) template. The frequency of priming in the self-primed M13 (C) system is not known, but is likely to be infrequent, certainly as compared with the activated DNA system. The results (Fig. 1) clearly show that adding the DNA-binding protein to the polymerase stimulated very markedly the polymerase activity in the M13 DNA assay systems, but stimulated the activity only very slightly when activated calf-thymus DNA was the template-primer. There was some variation between different preparations of DNAbinding protein in the degree of stimulation of polymerase on M13 DNA templates as indicated by the relatively large standard error bars with primer-template systems B and C (Fig. 1). However, the ability to stimulate
Fig. 1. (a) Effect on DNA polymerase activity of adding primerrecognition protein (1.6 mg) to DNA polymerase-a (16 mg) prior to the polymerase assay. The activity of the polymerase plus primerrecognition protein is expressed relative to the activity with polymerase alone. The latter was taken as 100%. Data were expressed this way in order to include polymerase preparations with different speci®c activities. The bars represent standard errors (n7). Three different template-primer systems were used. A, activated calf-thymus DNA; B, M13 DNA primed with a single sequencing primer; C, M13 DNA selfprimed by the primase associated with the polymerase; non-radioactive ribonucleoside triphosphates were included in this assay (Bryant et al., 1992). The activities in the controls (i.e. 100% levels) were for A, between 500 and 800 pmol dTMP incorporated mg 1 protein h 1 in different preparations and for B and C, between 45 and 100 pmol dTMP incorporated mg 1 protein h 1 in different preparations. (b) An individual DNA polymerase time-course with primed M13 DNA in the absence (A) or presence (B) of the primer-recognition protein (1.6 mg primer-recognition protein added to 16 mg polymerase). Bars represent SEM (n4).
polymerase activity on these infrequently primed templates was a consistent feature of the DNA-binding protein even though the extent of this ability varied. The demonstration that the DNA-binding protein stimulated DNA polymerase activity when added back to DNA-polymerase-a-primase raised the possibility that the binding protein is a processivity factor. However, the nascent DNA molecules synthesized in the presence of the DNA-binding protein were no longer than those synthesized in its absence (results not shown).
Discussion The results presented in this paper extend the earlier observations made on the 42 kDa DNA-binding protein associated with the DNA-polymerase-a-primase complex in pea. In particular, it has been shown that it stimulates
Primer-recognition protein in pea
the activity of DNA polymerase-a when the polymerase is working on infrequently primed or on self-primed templates and that this stimulation is not due to the protein acting as a processivity factor. Indeed, if it were a processivity factor it might be expected also to stimulate polymerase activity when activated calf-thymus DNA is used as a template-primer system. It is clear that with the latter system there was only a very slight stimulation of polymerase activity. This slight stimulation cannot be ascribed to the failure of the protein to bind to activated DNA because this particular template-primer system is an excellent `substrate' for binding. All these features lead us to suggest that the 42 kDa DNA-binding protein associated with DNA-polymerase-a-primase in pea is a primer-recognition protein, as tentatively suggested in the previous publication (Burton et al., 1997). The speci®c biochemical role of such a protein in DNA replication would be to prevent the DNA polymerase binding unproductively to the template strand, i.e. at a distance from the newly-synthesized primer (cf. Pritchard et al., 1983). This is ®rst time that such a protein has been described from plants. However, there has been a small number of reports of a similar protein from mammalian, particularly human, cells (Pritchard et al., 1983; Jindal and Vishwanatha, 1990). The human primer-recognition protein also has a molecular weight of 42 kDa and, rather surprisingly, appears to be identical to the glycolytic enzyme phosphoglycerate kinase (PGK). Furthermore, it has recently been shown that mammalian DNA polymerase-a is stimulated by puri®ed PGK in a manner consistent with it acting as a primer-recognition protein (Popanda et al., 1998). Preliminary evidence has been obtained in this laboratory that pea primer-recognition protein is at least immunologically similar to PGK (Bryant and Anderson, 1999) and a PGK-like antigen has been detected in pea leaf nuclei (Anderson et al., 1995). The possession of two completely different biochemical activities by one polypeptide has been now described for a small but growing number of examples for which the term `moonlighting proteins' has been coined (Jeffery, 1999). Current work is aimed at establishing whether pea primer-recognition protein is indeed the same protein as PGK.
Acknowledgements JAB thanks the Exeter University Research Committee and the Biotechnology and Biological Sciences Research Council (C&M 07236) and LEA thanks the National Science Foundation and the University of Illinois-Chicago Research Board for research grants. Collaboration between the two laboratories was funded by a NATO collaborative research grant. We thank Gill Davies for help in preparing the ®gure.
References Al-Rashdi J, Bryant JA. 1994. Puri®cation of a DNA-binding protein from a multi-protein complex associated with DNA polymerase-a in pea (Pisum sativum). Journal of Experimental Botany 45, 1867±1871. Anderson LE, Wang XW, Gibbons JT. 1995. Three enzymes of carbon metabolism or their antigenic analogues in pea leaf nuclei. Plant Physiology 108, 659±667. Bryant JA, Anderson LE. 1999. What's a nice enzyme like you doing in a place like this? A possible link between glycolysis and DNA replication. In: Bryant JA, Burrell MM, Kruger NJ, eds. Plant carbohydrate biochemistry. Oxford: Bios Scienti®c Publishers, 295±304. Bryant JA, Fitchett PN, Hughes SG, Sibson DR. 1992. DNA polymerase-a in pea is part of a large multi-protein complex. Journal of Experimental Botany 43, 31±45. Burton SK, Van't Hof J, Bryant JA. 1997. Novel DNA-binding characteristics of a protein associated with DNA polymerase-a in pea. The Plant Journal 12, 357±365. Daniel PP, Bryant JA, Barker DG. 1985. DNA ligase activity in pea (Pisum sativum) seedlings: development of a sensitive assay system and partial characterization of soluble and chromatinbound ligases. Biochemistry International 11, 645±652. Jeffery CJ. 1999. Moonlighting proteins. Trends in Biochemical Science 24, 8±11. Jindal HK, Vishwanatha JK. 1990. Puri®cation and characterization of primer-recognition proteins from HeLa cells. Biochemistry 29, 4767±4773. Popanda O, Fox G, Theilmann HW. 1998. Modulation of DNA polymerases a, d and e by lactate dehydrogenase and 3-phosphoglycerate kinase. Biochimica et Biophysica Acta 1397, 102±117. Pritchard CG, Weavers DT, Baril EF, De Pamphilis ML. 1983. DNA polymerase-a cofactors C1, C2 function as primerrecognition proteins. Journal of Biological Chemistry 258, 9810±9819. Waga S, Stillman B. 1994. Anatomy of a DNA replication fork revealed by reconstitution of SV40 DNA replication in vitro. Nature 369, 207±212. Waga S, Stillman B. 1998. The DNA replication fork in eukaryotic cells. Annual Review of Biochemistry 67, 721±751.